Subclass for random search tuning.
The random points are sampled by
In order to support general termination criteria and parallelization,
we evaluate points in a batch-fashion of size
Larger batches mean we can parallelize more, smaller batches imply a more fine-grained checking
of termination criteria.
TunerRandomSearch$new(batch_size = 1L) tnr("random_search")
Maximum number of configurations to try in a batch.
# see ?Tuner